Synthesis of natural-like acylphloroglucinols with anti-proliferative, anti-oxidative and tube-formation inhibitory activity

Article abstract of DOI:10.1016/j.ejmech.2014.08.017

Two series of natural and natural-like mono- and bicyclic acylphloroglucinols derived from secondary metabolites in the genus Hypericum (Hypericaceae) were synthesised and tested in vitro for anti-proliferative and tube-formation inhibitory activity in human microvascular endothelial cells (HMEC-1). In addition, their anti-oxidative activity was determined via an ORAC-assay. The first series of compounds (4a-e) consisted of geranylated monocyclic acylphloroglucinols with varying aliphatic acyl substitution patterns, which were subsequently cyclised to the corresponding 2-methyl-2-prenylchromane derivatives (5a and 5d). The second series involved compounds containing a 2,2-dimethylchromane skeleton with differing aromatic acyl substitution (6a-d and 7a-e). Compound 7a, (5,7-dihydroxy-2,2- dimethylchroman-6-yl)-(3,4-dihydroxyphenyl)methanone), showed the highest in vitro anti-proliferative activity with an IC₅₀ of 0.88 ± 0.08 μM and a remarkable anti-oxidative activity of 2.8 ± 0.1 TE from the ORAC test. Interestingly, the high anti-proliferative activity of these acylphloroglucinols was not associated with tube-formation inhibition. Compounds (E)-1-(3-(3,7-dimethylocta-2,6-dien-1-yl)-2,4,6-trihydroxyphenyl)-2- methylbutan-1-one (4d) and (5,7-dihydroxy-2,2-dimethylchroman-6-yl)(3,4- dimethoxyphenyl)methanone (6a) exhibited moderate to weak anti-proliferative effects (IC₅₀ 11.0 ± 1 μM and 48.0 ± 4.3 μM, respectively) and inhibited the capillary-like tube formation of HMEC-1 in vitro, whereas 7a was inactive. The most active compound in the ORAC assay was 7c, which exhibited an anti-oxidative effect of 6.6 ± 1.0 TE. However, this compound showed only weak activity during the proliferation assay (IC ₅₀ 53.8 ± 0.3) and did not inhibit tube-formation.

Full text of DOI:10.1016/j.ejmech.2014.08.017

European Journal of Medicinal Chemistry 85 (2014) 621e628
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Synthesis of natural-like acylphloroglucinols with anti-proliferative,
anti-oxidative and tube-formation inhibitory activity
a
b
b
b,
*
a, *
€
€
Qiu Sun , Sebastian Schmidt , Martina Tremmel , Jorg Heilmann , Burkhard Konig
a
€
€
Universitat Regensburg, Institute of Organic Chemistry, Universitatsstr. 31, D-93053 Regensburg, Germany
Universitat Regensburg, Pharmaceutical Biology, Universitatsstr. 31, D-93053 Regensburg, Germany
b
€
€
a r t i c l e i n f o
a b s t r a c t
Article history:
Two series of natural and natural-like mono- and bicyclic acylphloroglucinols derived from secondary
metabolites in the genus Hypericum (Hypericaceae) were synthesised and tested in vitro for anti-
proliferative and tube-formation inhibitory activity in human microvascular endothelial cells (HMEC-
1). In addition, their anti-oxidative activity was determined via an ORAC-assay. The first series of com-
pounds (4aee) consisted of geranylated monocyclic acylphloroglucinols with varying aliphatic acyl
substitution patterns, which were subsequently cyclised to the corresponding 2-methyl-2-
prenylchromane derivatives (5a and 5d). The second series involved compounds containing a 2,2-
dimethylchromane skeleton with differing aromatic acyl substitution (6aed and 7aee). Compound 7a,
(5,7-dihydroxy-2,2-dimethylchroman-6-yl)-(3,4-dihydroxyphenyl)methanone), showed the highest
Received 24 May 2014
Received in revised form
5 August 2014
Accepted 6 August 2014
Available online 7 August 2014
Keywords:
Acylphloroglucinol
Chromane
in vitro anti-proliferative activity with an IC50 of 0.88 0.08 mM and a remarkable anti-oxidative activity
HMEC-1
of 2.8 0.1 TE from the ORAC test. Interestingly, the high anti-proliferative activity of these acylphlor-
oglucinols was not associated with tube-formation inhibition. Compounds (E)-1-(3-(3,7-dimethylocta-
Anti-proliferative activity
Tube formation
ORAC
2,6-dien-1-yl)-2,4,6-trihydroxyphenyl)-2-methylbutan-1-one
dimethylchroman-6-yl)(3,4-dimethoxyphenyl)methanone (6a) exhibited moderate to weak anti-
proliferative effects (IC50 11.0 M and 48.0 4.3 M, respectively) and inhibited the capillary-like
(4d)
and
(5,7-dihydroxy-2,2-
1
m
m
tube formation of HMEC-1 in vitro, whereas 7a was inactive. The most active compound in the ORAC
assay was 7c, which exhibited an anti-oxidative effect of 6.6 1.0 TE. However, this compound showed
only weak activity during the proliferation assay (IC50 53.8 0.3) and did not inhibit tube-formation.
© 2014 Elsevier Masson SAS. All rights reserved.
1. Introduction
geranyl moieties. Both substituents can cyclise and oxidise, which
results in bicyclic, tricyclic or even more complex compounds [3].
Acylphloroglucinols are typical secondary metabolites biosyn-
thetically derived from the polyketide pathway that accumulate in
Hypericaceae [1] and Clusiaceae. Their biosynthesis begins with
three malonyl-CoA (coenzyme A) molecules reacting with an acti-
vated acyl-CoA to form an intermediate polyketide. A Claisen-like
reaction enzymatically cyclises this intermediate to an acylphlor-
oglucinol [2]. Depending on the starting material, the acyl moiety of
a natural acylphloroglucinol can be aliphatic, as observed in the
Genus Hypericum, or aromatic, as in the Genus Garcinia. The
structural diversity of acylphloroglucinols results mainly from
substitutions on the phloroglucinol core with several prenyl or
Acylphloroglucinols have attracted attention due to their broad
pharmacological activity. Some of these compounds have been
reported to exhibit anti-bacterial [4], cytotoxic [5], anti-oxidative
[6] and anti-depressant [7] effects. Hyperforin, likely the most
prominent acylphloroglucinol derivative, has been recently re-
ported to exhibit strong anti-proliferative effects [8] and inhibit
angiogenesis in vivo. Several key steps of this anti-angiogenic effect
have been determined pharmacologically in vitro [9]. Due to its
instability in aqueous solutions [10], complex structure and limited
availability, hyperforin is neither a drug candidate nor a good model
compound. Thus, synthesising pharmacologically active acyl-
phloroglucinol derivatives with higher stability and better solubi-
lity is an interesting challenge.
Recently, monocyclic and bicyclic acylphloroglucinol derivatives
with anti-proliferative activity for a human microvascular endo-
thelial cell line (HMEC-1) were isolated from Hypericum empe-
trifolium (SI: Fig. 1S) [11] [12], a plant commonly used in traditional
* Corresponding authors.
E-mail addresses: joerg.heilmann@chemie.uni-regensburg.de (J. Heilmann),
burkhard.koenig@chemie.uni-regensburg.de (B. Konig).
€
http://dx.doi.org/10.1016/j.ejmech.2014.08.017
0223-5234/© 2014 Elsevier Masson SAS. All rights reserved.

622
Q. Sun et al. / European Journal of Medicinal Chemistry 85 (2014) 621e628
Greek medicine. Structurally simpler than hyperforin, the natural
monocyclic compounds 4c, 4d and 5d exhibited potent anti-
proliferative activity at low micromolar concentrations (Table 1).
Due to the limited availability of the compound in plant materials,
our first step was to synthesise geranylated acylphloroglucinol, 4d,
from 3d to enable further in vitro pharmacological testing. Because
non-geranylated monocyclic acylphloroglucinols (3aee) were
inactive in the proliferation assay (data not shown), the remarkable
activity of the geranylated monocyclic compound, 4d, encouraged
us to more systematically investigate this class of molecules. First,
the acyl group length was varied, and the geranyl moiety was
subsequently cyclised, which provided the first series of gerany-
lated monocyclic and bicyclic aliphatic acylphloroglucinols 4aee,
5a and 5d (Scheme 1). Of the 2-methyl-2-prenyl-8yl chromane
derivatives obtained, compound 5d had been previously isolated
from H. empetrifolium [12] and H. amblycalyx [13], and 5a is a
natural-like analogue with a shortened acyl side chain.
Purifying the mixed compounds was difficult because geranyl
bromide reacts non-selective with the hydroxide groups or aro-
matic ring to yield over three side products. Finally, catalysing the
cyclisation using para-toluenesulfonic acid (pTSA) afforded the
target compounds 5a and 5d in 53 and 65% yield, respectively. The
2D NMR analysis of the products and a comparison of their spec-
troscopic data to literature values revealed they selectively cyclised
at the 2-OH position not the 4-OH position, which was unambig-
uously determined by comparing the 1D NMR data to literature.
To synthesise the second series, Amberlyst 15 was used to effi-
ciently catalyse the condensation of 1,3,5-trihydroxybenzene (1)
with isoprene (2) in 53% yield. The subsequent FriedeleCrafts
acylation yielded intermediates 6aed, which were subsequently
demethylated with BBr3 to form 7aee with 48e78% yield. Com-
pound 5e was prepared as an aliphatic 2,2-dimethyl-chroman-6yl
analogue of the 2-methyl-2-prenyl-chroman-8yl derivative 5d.
A second series of compound syntheses (Scheme 2) was inspired
by the aromatic acyl moieties in several natural acylphloroglucinols
with chromane skeletons [14]. The bicyclic acylphloroglucinols
obtained (6aed, 7aee) bear aromatic acyl groups at C-6 in the
chromane skeleton with different hydroxyl or methoxy substitu-
tion patterns. Several pharmacologically active natural acylphlor-
oglucinols containing aromatic substitutions, such as rottlerin and
its derivatives [15] or (ꢀ)-3-deoxy-MS-II [16], have a 2,2-dimethyl-
chromane core. Therefore, we used this skeleton to synthesise
6aed and 7aee instead of 2-methyl-2-prenyl-chromane, which
simplifies the structure.
2.2. Biological evaluation
The anti-proliferative activity of monocyclic aliphatic acyl-
phloroglucinols (4aee) in HMEC-1 increased with increasing logP
(Table 1). The IC50 values decreased from 22.5
8.7 mM (4a) to
8.7 3.2 M (4e) upon extending the acyl-side chain. The anti-
m
proliferative activity of the bicyclic aliphatic acylphloroglucinols
(5a and 5d) had the same range, and their activity likewise
increased with the logP to yield IC50 values of 20.3
5.2 and
13.3 3.8 M, respectively. Increasing the number of carbon atoms
m
in the acyl group provides higher lipophilicity, which allows the
compound to better penetrate and cross cellular membranes.
Accordingly, changing the 2-prenyl (in 5d) to a methyl group (in 5e)
decreased the lipophilicity (5d logP 5.5 versus 5e logP 3.77) and
2. Results and discussion
anti-proliferative effect (5d 13.3 3.8 and 5e 29.3 5.2 mM). The
2.1. Synthesis
reduced anti-proliferative effect for 5e matches the slight activity
reduction for aliphatic chromanes acylated at C-6 relative to the C-8
acyl derivatives, as reported for natural acylphloroglucinols by
Schmidt et al. [11].
Compounds 3aee were synthesised via a FriedeleCrafts acyla-
tion (Scheme 1) with a 55e81% yield. Treating 3aee and geranyl
bromide with anhydrous potassium carbonate in acetone yielded
the alkylation products 4aee with moderate yields from 55 to 60%.
Except for compounds 7a and 7e, which exhibited a strong anti-
proliferative activity with IC50 values of 0.88
0.08 and
7.6 1.5 M, respectively, acylphloroglucinols in the second series
m
Table 1
with an aromatic acyl side chain and 2,2-dimethyl-chromane-6yl
core exhibited lower anti-proliferative activities relative to their
aliphatic 8 yl analogues (Table 1) and the 6 yl derivative 5e. The
strong activity for 7a resulted from its 3,4-dihydroxy-phenyl sub-
stitution, whereas protecting the hydroxyl groups with methyl
groups or changing the hydroxyl substitution pattern dramatically
decreased the activity. Chromane 7a is z50-fold more active than
Anti-proliferative activity of aliphatic monocyclic (4aee), bicyclic (5a, 5d and 5e)
acylphloroglucinols and aromatic bicyclic (6ae7e) acylphloroglucinols on HMEC-1
(mean value
(Eh). Trolox equivalents (TE) were determined in an ORAC assay (1e5
acylphloroglucinols with the exception of 6d.
SD, n ¼ 3), status, calculated values for logP and redox-potential
mM) for all
Type
Compound
IC50
[m
M] LogPa Trolox
equivalents
(TE)
Redox
potential [V]b
the 3,4-dimethoxy-benzoyl 6a (48.0 4.3
mM) and z30-fold more
active than the 3,5-dihydroxybenzoyl derivative 7b (25.5 3.4
m
M).
Aliphatic side 4a
22.5 8.7 4.62 1.0 0.1
22.4 3.7 5.13 0.6 0.1
21.0 0.4 5.49 0.8 0.1
11.0 1.2 6.00 0.4 0.1
8.7 3.2 6.50 0.3 0.1
20.3 5.2 4.22 0.8 0.1
13.3 3.8 5.59 0.3 0.1
29.3 5.2 3.77 1.2 0.2
48.0 4.3 4.94 1.3 0.1
43.4 15.3 5.54 1.1 0.1
51.5 3.2 4.42
0.88 0.08 3.68 2.8 0.1
25.5 3.4 5.30 3.0 0.1
53.8 0.3 5.07 6.6 1.0
23.9 2.4 3.66 3.0 0.4
7.6 1.5 4.53 2.1 0.2
1.06
0.86
0.86
0.86
0.86
1.23
1.23
0.92
0.98
0.96
0.95
0.80
0.87
1.10
0.75
0.67
chain
4b
4c
4d
4e
5a
5d
5e
Furthermore, dimethoxylation significantly reduced the water
solubility and hampered using aqueous solutions in vitro. Thus,
compound 6b, which bears a 3, 5-dimethoxy-benzoyl group was
not tested further in our assays. In contrast to the aliphatic series,
the logPs for compounds 6aed (without 6b) and 7aee did not
correlate to their anti-proliferative activity in vitro, which implies
they may act via a different mechanism than acylphloroglucinols
4aee, especially for the highly active compounds 7a and 7e.
The anti-oxidative activity of the acylphloroglucinols was tested
via an ORAC (Oxygen Radical Absorbance Capacity) assay. Phlor-
oglucinols containing aromatic acyl groups generally showed
higher anti-oxidative effects (1.1 0.1e6.6 1.0 TE) relative to their
monocyclic and bicyclic counterparts with aliphatic acyl substitu-
tion (0.3 0.1e1.2 0.2 TE). We also determined the ORAC activity
of natural 8-hydroxy-6-yl chromanes, 1-[5,7-dihydroxy-2-methyl-
2-(4-methylpent-3-enyl)-chroman-6-yl]-2-methylpropan-1-one
Aromatic side 6a
chain
6c
6d
7a
7b
7c
7d
7e
Xanthohumol 11.4 1.1
Trolox
2.3 0.2 [30]
0.49
a
The partition coefficient (logP) of compounds were calculated with software
ACD (Advanced Chemistry Development)/Lab 12.0.
b
Determined via cyclic voltammetry.
and
1-[5,7-dihydroxy-2-methyl-2-(4-methylpent-3-enyl)-

Q. Sun et al. / European Journal of Medicinal Chemistry 85 (2014) 621e628
623
Scheme 1. Synthesis of compounds 4aee and 5a, 5d; Reagents and conditions: (a) acyl chloride, AlCl3, CS2-PhNO2, 55 ꢁC, 2 h; (b) geranyl bromide, K2CO3, acetone, reflux overnight;
(c) pTSA, benzene, reflux, 2 h.
Scheme 2. Synthesis of compounds 5e, 6aed and 7aee; reagents and conditions: (a) isoprene, Amberlyst 15, THF/hexane; (b) benzoyl chloride, AlCl3, CH2Cl2, ꢀ5 ꢁC to r.t. (room
temperature), overnight; (c) BBr3, CH2Cl2, ꢀ78 ꢁC to r.t., overnight; (d) AlCl3, CH2Cl2, r.t.
chroman-6-yl]-2-methylbutan-1-one, which were recently isolated
from H. empetrifolium [11], relative to the aliphatic 6-hydroxy-8-yl
substitutions typically exhibited lower redox-potentials than the
aliphatic 8-yl derivatives.
5a. Their activities were 0.3
0.1 and 0.4
0.1 TE, which indi-
Interestingly, the high anti-proliferative activity of acylphlor-
oglucinols was not associated with tube-formation inhibition.
Compound 7a, which showed the strongest activity during the
proliferative assay, was inactive during the tube-formation assay.
Only compounds 4d and 6a maintained their activity across the
same concentration range in the proliferative assay in both series.
Thus, the less specific anti-proliferative effect, which is potentially
connected to a nonspecific cytotoxic activity, can be decoupled
from the more specific tube-formation inhibition. Tube-formation
cated that the anti-oxidative activity of the aliphatic 8-hydroxy-6-
yl chromanes was reduced further relative to the 6-hydroxy-8-yl
chromanes in our series. The strongest activity was observed for
compound 7c with 6.6 1.0 TE, which is comparable to strong anti-
oxidants such as caffeic acid or protocatechuic acid as reported by
Davalos et al. [17]. A high redox potential and radical scavenger
activity were correlated for several phenolic compounds [18], we
also determined the redox potential for the acylphloroglucinols
using cyclic voltammetry measurements with Trolox as the refer-
ence compound. However, there was no correlation between anti-
oxidative activity measured in the ORAC assay and redox potential
within the tested acylphloroglucinols. The aliphatic 6-hydroxy-8-yl
chromanes 5a and 5d (Eh ¼ 1.23 V each) were unable to reach the
lower redox potential of the Trolox reference compound
(Eh ¼ 0.49 V), and surprisingly, the 6-acyl chromanes with aromatic
is inhibited by compound 4d at 50, 25, 12.5 and 6.25
mM (Fig. 1)
and by compound 6a at 100, 50, 25 and 12.5 M (SI: Fig. 2S). Un-
m
treated wells and inactive acylphloroglucinols exhibited branching
and interconnectivity, but endothelial cells treated with 4d and 6a
were dispersed and formed small cell clumps without significant
network formation. The effect of 4d was superior to that of 6a at the
same concentration.

624
Q. Sun et al. / European Journal of Medicinal Chemistry 85 (2014) 621e628
Fig. 1. Tube-formation assay using compound 4d in different concentrations: 1. Negative control (cells untreated), 2. 50 mM, 3. 25 mM, 4. 12.5 mM, 5. 6.25 mM.
3. Conclusions
on a Biotage® Isolera™ Spektra One device. Silica gel 60 M
(40e63 m, Merck) was used for the flash column chromatography.
m
Overall, the results suggest acylphloroglucinols containing
simpler substitution patterns than hyperforin can show strong anti-
proliferative effects and remarkable tube formation inhibition,
which are not correlated. For monocyclic and bicyclic compounds
containing aliphatic side-chains, the anti-proliferative effect is
influenced by the side chain length, prenyl substitution and
increasing lipophilicity. A simple correlation to logP was not
observed for acylphloroglucinol with an aromatic acyl substitution.
The prerequisite for a strong activity during the ORAC assay was the
presence of an aromatic acyl-substituent especially one with a 3,4-
dihydroxy-substitution. The ORAC assay activity did not correlate to
the cyclic voltammetry redox potential.
The recently described biological importance of different
terpene substituents at the hydroxyl group monocyclic acylphlor-
oglucinols [11] makes the systematic synthesis of a library of O-
substituted mono- and bicyclic acylphloroglucinols interesting.
Furthermore, investigating the pharmacological activity of mono-
cyclic phloroglucinols with aromatic acyl moieties is in progress.
The starting materials and reagents were purchased from com-
mercial suppliers and used without further purification. The sol-
vents were p.a. grade for the reaction mixtures and industrial grade
for the flash column chromatography. Analytical TLC was per-
formed on silica gel coated alumina plates (MN TLC sheets ALU-
GRAM® Xtra SIL G/UV254). The visualisation was performed using
UV-light (254 and 366 nm). The logP values were calculated using
ACD/Lab 12.0 software. The redox potential experiments were
performed using an Autolab PGSTAT3202N to determine the Eh (in
V).
The purity of the tested compounds was determined via
analytical HPLC, Elite LaChrom (VWR, Darmstadt) using EZChro-
mElite 3.1.7 Software to measure the purity in percent. Column:
Hibar® 2504 Purospher Star RP18e (5
mm), wavelength: peak
maximum, gradient: FA 0.1%eMeCN 95%, 5 / 100 (MeCN 95%)
within 30 min before washing and equilibrating the column to the
starting conditions for a further 10 min, flow: 0.750 mL/min. The
concentration of the tested compounds was 100
mM, and the in-
jection volume was 20
higher.
ml. The compound purities were 90% or
4. Experimental
1H, 13C and 2D NMR spectra were obtained at 298 K using a
Bruker AVANCE 300 spectrometer (operating at 300.13 MHz for 1H
and 75.47 MHz for 13C), Bruker AVANCE 400 spectrometer (oper-
ating at 400.13 MHz for 1H and 100.62 MHz for 13C) and Bruker
AVANCE 600 spectrometer (operating at 600.25 MHz for 1H and
150.93 MHz for 13C) (Bruker, Karlsruhe, Germany). The spectra were
obtained using chloroform-d (99.8%, Deutero GmbH) or methanol-
d4 (99.8%, Deutero GmbH) and referenced against non-deuterated
(1H)/deuterated (13C) solvents. The shift values (dH and dC) are al-
ways given in ppm with J values in Hz. The melting points were
measured using a Stanford Research Systems OptiMelt MPA 100.
The high-resolution mass spectra were obtained using a Finnigan
MAT SSQ 710 A spectrometer at 70 eV (HREIMS, positive and
negative mode) or an Agilent 6540 UHD (HRESIMS, positive and
negative mode). Automated flash chromatography was performed
4.1. 2,2-Dimethyl-5,7-dihydroxychromane (2)
The sulfonic acid cation-exchange resin Amberlyst 15 (10.0 g)
was stirred in dry THF (tetrahydrofurane,120 mL) while refluxing at
65ꢀ75 ꢁC. Then, 1,3,5-trihydroxybenzene (5.0 g) was added to the
resin and followed with 3.0 g of isoprene in hexane (40 mL) over
2 h. The reaction mixture was stirred for 1 h under reflux before
removing the heat source and adding 100 mL of Et2O. The resin was
filtered via vacuum filtration and rinsed with acetone (75 mL). The
crude product mixture was purified via flash chromatography (PE
(bp. 50ꢀ70 ꢁC)/EtOAc 1:1) to afford compound 2 as white solid.
Yield: 53%, 4.08 g. Mp. 156ꢀ157 ꢁC (162 ꢁC [19]); 1H NMR (300 MHz,
MeOD)
d
5.86 (d, J ¼ 2.4 Hz, 1H), 5.73 (d, J ¼ 2.3 Hz, 1H), 2.52 (t,
J ¼ 6.8 Hz, 2H), 1.71 (t, J ¼ 6.8 Hz, 2H),1.26 (s, 6H). 13C NMR (75 MHz,
MeOD)
d 157.5, 157.3, 156.5, 101.5, 96.4, 95.6, 74.9, 33.7, 27.0, 17.7.

Q. Sun et al. / European Journal of Medicinal Chemistry 85 (2014) 621e628
625
These spectroscopic data are in accordance with the literature
4.3.2. (E)-1-(3-(3,7-Dimethylocta-2,6-dien-1-yl)-2,4,6-
values [20].
trihydroxyphenyl)propan-1-one (4b)
Yield: 182 mg, 60%. Pale yellow solid. Mp. 114ꢀ116 ꢁC. 1H NMR
4.2. General procedure for the synthesis of 3aee
(400 MHz, CDCl3) d 11.40 (s, br, 1H), 8.60 (bs, 1H), 6.01 (s, 1H), 5.84
(s, 1H), 5.25 (td, J ¼ 7.2, 1.1 Hz, 1H), 5.05 (m, 1H), 3.38 (d, J ¼ 7.2 Hz,
AlCl3 (4 equiv) was added to a stirred phloroglucinol suspension
(1 equiv) in CS2. Nitrobenzene was added to the solution over
30 min. The solution was refluxed at 55 ꢁC for 30 min. Acyl chloride
(1 equiv) dissolved in nitrobenzene was added to the reaction
mixture over 30 min before heating for another 30 min. The reac-
tion mixture was allowed to cool with stirring and then poured into
an ice-water bath. Afterwards, 3 M HCl was added. The organic
solvents were removed under a reduced pressure, and the oily
residue containing the acylphloroglucinol was extracted with Et2O.
After removing the Et2O, the crude product was purified via flash
chromatography (PE (bp. 50ꢀ70 ꢁC)/EtOAc 5:1 / 3:1) [21].
2H), 3.09 (q, J ¼ 7.2 Hz, 2H), 2.17e2.04 (m, 4H), 1.81 (d, J ¼ 0.7 Hz,
3H), 1.68 (d, J ¼ 0.6 Hz, 3H), 1.60 (s, 3H), 1.17 (t, J ¼ 7.2 Hz, 3H). 13
C
NMR (101 MHz, CDCl3)
d 206.8, 162.3, 160.7, 160.1, 140.2, 132.2,
123.6, 121.4, 105.5, 104.9, 95.4, 39.7, 37.3, 26.3, 25.7, 21.6, 17.7, 16.2,
8.6. HRMS (ESI) calcd for C19H26O4 [MꢀH]ꢀ 317.1758 found
317.1752.
4.3.3. (E)-1-(3-(3,7-Dimethylocta-2,6-dien-1-yl)-2,4,6-
trihydroxyphenyl)-2-methylpropan-1-one (4c)
Yield: 219 mg, 55%. Pale yellow solid. Mp. 109ꢀ111 ꢁC. 1H NMR
(400 MHz, CDCl3) d 11.55 (s, br, 1H), 8.35 (s, br, 1H), 5.93 (s, 1H), 5.83
(s, 1H), 5.32e5.20 (m, 1H), 5.12e4.97 (m, 1H), 3.92e3.84 (m, 1H),
3.38 (d, J ¼ 6.9 Hz, 2H), 2.10 (t, J ¼ 5.0 Hz, 4H), 1.81 (s, 3H), 1.69 (s,
3H), 1.60 (s, 3H), 1.18 (d, J ¼ 6.7 Hz, 6H). 13C NMR (101 MHz, CDCl3)
4.2.1. 1-(2,4,6-Trihydroxyphenyl)ethanone (3a)
Yield: 1.89 g, 75%. Yellow solid. Mp. 219ꢀ221 ꢁC (216ꢀ218 ꢁC
[22]); 1H NMR (300 MHz, MeOD)
d
5.79 (s, 2H), 2.59 (s, 3H). The
d 210.6, 162.6, 160.7, 159.8, 140.2, 132.2, 123.6, 121.4, 105.6, 104.2,
spectroscopic data are in accordance with the literature values [23].
95.5, 39.7, 39.3, 26.3, 25.7, 21.6, 19.3, 19.3, 17.7, 16.2. HRMS (ESI)
calcd for C20H28O4 [MꢀH]ꢀ 331.1915 found 331.1906.
4.2.2. 1-(2,4,6-Trihydroxyphenyl)propan-1-one (3b)
Yield: 826 mg, 75%. Pale yellow solid. Mp. 175ꢀ176 ꢁC (174 ꢁC
4.3.4. (E)-1-(3-(3,7-Dimethylocta-2,6-dien-1-yl)-2,4,6-
trihydroxyphenyl)-2-methylbutan-1-one (4d)
[24]); 1H NMR (300 MHz, MeOD)
d
5.79 (s, 2H), 3.06 (q, J ¼ 7.3 Hz,
2H), 1.12 (t, J ¼ 7.3 Hz, 3H). The spectroscopic data are in accordance
Yield: 214 mg, 60%. Pale yellow solid. Mp. 96ꢀ97 ꢁC. 1H NMR
with the literature values [23].
(300 MHz, CDCl3) d 11.67 (s, br, 1H), 8.20 (s, br, 1H), 5.90 (s, 1H), 5.78
(s, 1H), 5.21 (t, J ¼ 6.0 Hz, 1H), 5.11e4.91 (m, 1H), 3.75e3.64 (m, 1H),
3.33 (d, J ¼ 7.1 Hz, 2H), 2.05 (s, br, 4H), 1.77 (s, 3H), 1.63 (s, 3H), 1.55
(s, 3H), 1.36 (tt, J ¼ 14.5, 7.3 Hz, 2H), 1.11 (d, J ¼ 6.8 Hz, 3H), 0.86 (t,
4.2.3. 2-Methyl-1-(2,4,6-trihydroxyphenyl)propan-1-one (3c)
Yield: 1.35 g, 81%. Pale yellow solid. Mp. 138ꢀ140 ꢁC (140 ꢁC
[24]); 1H NMR (300 MHz, MeOD)
d
5.80 (s, 2H), 3.97 (dt, J ¼ 13.5,
J ¼ 7.4 Hz, 3H). 13C NMR (75 MHz, CDCl3)
d 210.8, 162.7, 161.2, 160.3,
6.8 Hz, 1H), 1.12 (d, J ¼ 6.7 Hz, 6H). The spectroscopic data are in
139.4, 132.1, 123.7, 121.7, 105.7, 104.7, 95.4, 45.9, 39.7, 27.0, 26.3, 25.7,
21.6, 17.7, 16.7, 16.2, 12.0. HRMS (ESI) calcd for C21H30O4 [MþH]þ
347.2215 found 347.2217.
accordance with the literature values [25].
4.2.4. 2-Methyl-1-(2,4,6-trihydroxyphenyl)butan-1-one (3d)
Yield: 3.12 g, 55%. Pale yellow oil. 1H NMR (300 MHz, MeOD)
4.3.5. (E)-1-(3-(3,7-Dimethylocta-2,6-dien-1-yl)-2,4,6-
trihydroxyphenyl)-2-methylpentan-1-one (4e)
d
5.81 (s, 2H), 3.83 (sext, J ¼ 6.7 Hz,1H),1.89e1.67 (m,1H),1.47e1.26
(m, 1H), 1.09 (d, J ¼ 6.8 Hz, 3H), 0.87 (t, J ¼ 7.4 Hz, 3H). The spec-
Yield: 78 mg, 55%. Pale yellow solid. Mp. 77ꢀ79 ꢁC. 1H NMR
troscopic data are in accordance with the literature values [21].
(300 MHz, CDCl3)
d
5.84 (s, 1H), 5.26 (t, J ¼ 6.0 Hz, 1H), 5.14e4.91
(m, 1H), 3.90e3.79 (m, 1H), 3.37 (d, J ¼ 7.1 Hz, 2H), 2.15e2.03 (m,
4H), 1.81 (s, 3H), 1.67 (s, 3H), 1.59 (s, 3H), 1.42e1.24 (m, 4H), 1.15 (d,
J ¼ 6.8 Hz, 3H), 0.89 (t, J ¼ 6.0 Hz, 3H). 13C NMR (75 MHz, CDCl3)
4.2.5. 2-Methyl-1-(2,4,6-trihydroxyphenyl)pentan-1-one (3e)
Yield: 1.32 g, 60%. Pale yellow oil. 1H NMR (300 MHz, MeOD)
d
5.79 (s, 2H), 3.96 (dt, J ¼ 13.2, 6.6 Hz, 1H), 1.85e1.63 (m, 1H),
d 210.7, 162.8, 160.8, 159.6, 140.1, 132.2, 123.6, 121.5, 105.7, 104.6,
1.43e1.18 (m, 3H), 1.10 (d, J ¼ 6.7 Hz, 3H), 0.90 (t, J ¼ 6.0 Hz, 3H).
95.5, 44.2, 39.7, 36.2, 26.3, 25.7, 21.6, 20.7, 17.7, 17.2, 16.2, 14.2. HRMS
HRMS (ESI) calcd for C12H16O4 [MþH]þ 225.1121 found 225.1123.
(EI) calcd for C22H32O4 [MꢀH]ꢀ 359.2228 found 359.2229.
4.3. General procedure for synthesising 4aee
4.3.6. 1-(5,7-Dihydroxy-2-methyl-2-(4-methylpent-3-en-1-yl)
chroman-8-yl)ethanone (5a)
A mixture of 3aee (1 equiv), geranyl bromide (1 equiv), and
K2CO3 (2 equiv) in acetone was refluxed for 24 h. Evaporating the
acetone, adding a 2 N HCl solution, extracting with EtOAc, and
removing the solvent were followed by a flash column chroma-
tography on silica gel with PE (bp. 50ꢀ70 ꢁC)/EtOAc (5:1) to yield
the corresponding products [26].
Yield: 107 mg, 53%. Pale yellow oil. 1H NMR (300 MHz, CDCl3)
d
13.87 (s, 1H), 6.33 (s, 1H), 5.95 (s, 1H), 5.09 (ddd, J ¼ 7.1, 4.1, 1.2 Hz,
1H), 2.64 (s, 3H), 2.63e2.55 (m, 2H), 2.10e2.02 (m, 2H), 1.95e1.63
(m, 5H), 1.60 (s, 3H), 1.35 (s, 3H). 13C NMR (75 MHz, CDCl3)
203.6,
d
164.8, 160.7, 157.6, 132.2, 123.7, 106.3, 99.9, 95.2, 78.4, 39.6, 33.4,
29.4, 25.7, 24.2, 22.6, 17.6, 16.0. HRMS (ESI) calcd for C18H24O4
[MþH]þ 305.1747 found 305.1749.
4.3.1. (E)-1-(3-(3,7-Dimethylocta-2,6-dien-1-yl)-2,4,6-
trihydroxyphenyl)ethanone (4a)
4.3.7. 1-(5,7-Dihydroxy-2-methyl-2-(4-methylpent-3-en-1-yl)
chroman-8-yl)-2-methylbutan-1-one (5d)
Yield: 385 mg, 58%. Pale yellow solid. Mp. 112ꢀ114 ꢁC. 1H NMR
(300 MHz, CDCl3)
d
11.41 (s, br, 1H), 8.62 (s, br, 1H), 6.35 (s, 1H), 5.86
Yield: 33 mg, 65%. Pale yellow oil. 1H NMR (300 MHz, CDCl3)
(s, 1H), 5.25 (dd, J ¼ 7.2, 6.0 Hz, 1H), 5.13e4.97 (m, 1H), 3.37 (d,
d
14.12 (s, 1H), 5.97 (s, 1H), 5.09 (t, J ¼ 6.8, 1H), 3.75 (sext, J ¼ 6.8 Hz,
J ¼ 7.1 Hz, 2H), 2.67 (s, 3H), 2.14e2.04 (m, 4H), 1.81 (d, J ¼ 0.9 Hz,
1H), 2.58 (m, 2H), 2.10e2.03 (m, 2H), 1.87e1.77 (m, 3H), 1.70 (m,
2H), 1.68 (s, 3H), 1.60 (s, 3H), 1.40 (m, 1H), 1.36 (s, 3H), 1.14 (d,
J ¼ 6.8 Hz, 3H), 0.89 (t, J ¼ 7.4 Hz, 3H). 13C NMR (75 MHz, CDCl3)
3H), 1.67 (s, 3H), 1.59 (s, 3H). 13C NMR (75 MHz, CDCl3)
d 203.8,
162.4, 161.4, 160.4, 140.0, 132.2, 123.6, 121.4, 105.4, 105.2, 95.3, 39.7,
32.9, 26.3, 25.7, 21.5, 17.7, 16.2. HRMS (ESI) calcd for C18H24O4
[MþH]þ 305.1747 found 305.1745.
d
210.7, 165.0, 160.5, 157.0, 132.2, 123.6, 106.0, 100.1, 95.5, 78.4, 46.3,
39.7, 29.1, 27.0, 25.6, 24.0, 22.6, 17.5, 16.6, 16.2, 12.0. HRMS (ESI) calc.

626
Q. Sun et al. / European Journal of Medicinal Chemistry 85 (2014) 621e628
for C21H30O4 [MþH]þ 347.2217 found 347.2216. The spectroscopic
4.5. General procedure for synthesis of 7aee [28]
data are in accordance with the literature [11].
BBr3 (2.5 mmol) was added to a solution of 6aed (0.84 mmol) in
DCM (30 mL) at ꢀ78 ꢁC. The reaction mixture was stirred for
30 min, warmed to room temperature and stirred overnight under
an N2 atmosphere. The mixture was quenched by carefully pouring
into iced water. The aqueous layer was extracted three times with
EtOAc and washed with 5% NaHSO3 and water before drying the
organic layer over anhydrous Na2SO4 and removing the solvent
under reduced pressure to yield the target products.
4.4. General procedure for synthesis of 5e and 6aed [27]
Anhydrous AlCl3 (3 equiv) was added portionwise to a solution
of 2 (1 equiv) in CH2Cl2. The corresponding chloride (1 equiv) was
then added dropwise to keep the reaction temperature
below ꢀ5 ꢁC. After stirring overnight at room temperature, the
mixture was poured into ice-water and extracted with CH2Cl2. The
combined organic layer was successively washed with saturated
NaHCO3 and brine, and then dried over anhydrous Na2SO4. The
corresponding products were obtained after flash column chro-
matography (PE (bp. 50e70 ꢁC)/EtOAc 5:1 / 3:1).
4.5.1. 6-(3,4-Dihydroxy)benzyl-5,7-dihydroxy-2,2-dimethyl-3,4-
dihydro-2H-1-benzopyran (7a)
Yield: 93 mg, 78%. Pale yellow solid. Mp: 79e81 ꢁC. 1H NMR
(600 MHz, MeOD)
d
7.06 (d, J ¼ 2.0 Hz, 1H), 7.02 (dd, J ¼ 8.2, 2.0 Hz,
1H), 6.75 (d, J ¼ 8.2 Hz, 1H), 5.95 (s, 1H), 2.55 (s, 2H), 1.65 (t,
4.4.1. 1-(5,7-Dihydroxy-2,2-dimethylchroman-6-yl)-2-
methylbutan-1-one (5e)
J ¼ 6.8 Hz, 2H), 1.04 (s, 6H). 13C NMR (151 MHz, MeOD)
d 199.5,
Yield: 52 mg, 26%. Pale yellow solid. Mp. 145ꢀ146 ꢁC. 1H NMR
161.6, 160.7, 156.3, 150.6, 145.5, 134.4, 123.2, 117.2, 115.3, 107.6, 101.5,
95.3, 76.0, 33.1, 26.6, 17.6. HRMS (ESI) calcd for C18H18O6 [MþH]þ
331.1182 found 331.1181.
(300 MHz, CDCl3)
d 13.70 (s, br, 1H), 6.60 (s, br, 1H), 5.74 (s, 1H),
3.81e3.70 (m, 1H), 2.58 (t, J ¼ 6.8 Hz, 2H), 1.78 (t, J ¼ 6.8 Hz, 2H),
1.46e1.34 (m, 2H), 1.32 (s, 6H), 1.16 (d, J ¼ 6.7 Hz, 3H), 0.91 (t,
J ¼ 7.4 Hz, 3H). 13C NMR (75 MHz, CDCl3)
d 210.3, 164.1, 160.2, 157.7,
103.8, 101.7, 95.7, 76.0, 45.8, 32.1, 27.0, 26.7, 16.7, 16.1, 12.0.
4.5.2. 6-(3,5-Dihydroxy)benzyl-5,7-dihydroxy-2,2-dimethyl-3,4-
dihydro-2H-1-benzopyran (7b)
Yield: 38 mg, 53%. Pale yellow solid. Mp. 99ꢀ101 ꢁC. 1H NMR
(300 MHz, MeOD)
d
6.48 (d, J ¼ 2.2 Hz, 2H), 6.40e6.32 (m, 1H), 5.74
4.4.2. 6-(3,4-Dimethoxy)benzyl-5,7-dihydroxy-2,2-dimethyl-3,4-
dihydro-2H-1-benzopyran (6a)
(s, 1H), 2.59 (t, J ¼ 6.8 Hz, 2H), 1.80 (t, J ¼ 6.8 Hz, 2H), 1.33 (s, 6H). 13
C
NMR (75 MHz, MeOD) d 201.1, 162.7, 162.2, 162.2, 160.1, 160.1, 159.5,
Yield: 60 mg, 68%. White solid. Mp. 127ꢀ128 ꢁC. 1H NMR
159.0, 154.2, 107.5, 106.1, 105.9, 101.6, 96.6, 76.7, 33.2, 27.0, 17.2.
(600 MHz, CDCl3)
d
10.73 (s, br, 1H), 7.31 (dd, J ¼ 8.3, 2.0 Hz, 1H),
HRMS (ESI) calcd for C18H18O6 [MþH]þ 331.1182 found 331.1174.
7.23 (d, J ¼ 1.9 Hz, 1H), 6.94 (d, J ¼ 8.3 Hz, 1H), 5.90 (s, 1H), 3.94 (d,
J ¼ 14.3 Hz, 6H), 2.61 (t, J ¼ 6.8 Hz, 2H), 1.81 (t, J ¼ 6.8 Hz, 2H), 1.36
4.5.3. 6-(4-Hydroxy)benzyl-5,7-dihydroxy-2,2-dimethyl-3,4-
dihydro-2H-1-benzopyran (7c)
(s, 6H). 13C NMR (151 MHz, CDCl3)
d 196.0, 161.7, 160.9, 158.1, 152.9,
149.6, 131.9, 122.2, 110.9, 110.9, 103.9, 101.8, 96.9, 76.0, 56.2, 56.1,
32.1, 26.8, 16.3. HRMS (ESI) calcd for C20H22O6 [MꢀH]ꢀ 357.1345
found 357.1344.
Yield: 83 mg, 48%. Pale brown solid. Mp. 157ꢀ159 ꢁC. 1H NMR
(300 MHz, MeOD)
d 7.60e7.41 (m, 2H), 6.81e6.63 (m, 2H), 5.77 (s,
1H), 2.59 (t, J ¼ 6.8 Hz, 2H), 1.80 (t, J ¼ 6.8 Hz, 2H), 1.33 (s, 6H). 13
C
NMR (75 MHz, MeOD) d 199.4, 162.5, 161.8, 161.3, 159.2, 133.7, 132.7,
4.4.3. 6-(3,5-Dimethoxy)benzyl-5,7-dihydroxy-2,2-dimethyl-3,4-
dihydro-2H-1-benzopyran (6b)
115.3, 106.2, 101.8, 96.6, 76.5, 33.3, 27.0, 17.4. HRMS (ESI) calcd for
C
18H18O5 [MþH]þ 315.1232 found 315.1225.
Yield: 200 mg, 52%. Pale yellow solid. Mp. 145ꢀ147 ꢁC. 1H NMR
(600 MHz, CDCl3)
d
11.02 (s, 1H), 7.28 (s, br, 1H), 6.70 (d, J ¼ 2.3 Hz,
4.5.4. 6-(3,5-Dimethoxy-4-hydroxy)benzyl-5,7-dihydroxy-2,2-
dimethyl-3,4-dihydro-2H-1-benzopyran (7d)
2H), 6.62 (t, J ¼ 2.2 Hz, 1H), 5.87 (s, 1H), 3.83 (s, 6H), 2.60 (t,
J ¼ 6.8 Hz, 2H), 1.80 (t, J ¼ 6.8 Hz, 2H), 1.35 (s, 6H). 13C NMR
Yield: 68 mg, 43%. Pale yellow solid. Mp. 182ꢀ184 ꢁC. 1H NMR
(151 MHz, CDCl3)
d 196.7, 162.4, 161.6, 158.8, 141.9, 104.5, 104.1,
(300 MHz, CDCl3)
d 10.86 (s, 1H), 6.95 (s, 3H), 5.99 (s, 1H), 5.90 (s,
103.7, 101.8, 97.1, 76.2, 55.6, 32.0, 26.7, 16.1. HRMS (ESI) calcd for
1H), 3.91 (s, 6H), 2.61 (t, J ¼ 6.8 Hz, 2H), 1.80 (t, J ¼ 6.8 Hz, 2H), 1.35
C
20H22O6 [MþH]þ 359.1495 found 359.1491.
(s, 6H). 13C NMR (75 MHz, CDCl3)
d
195.9, 161.9, 161.0, 158.0, 147.4,
138.9, 130.3, 105.3, 103.8, 101.9, 97.0, 76.1, 56.6, 32.1, 26.8, 16.2.
HRMS (ESI) calcd for C20H22O7 [MþH]þ 375.1440 found 375.1438.
4.4.4. 6-(4-Methoxy)benzyl-5,7-dihydroxy-2,2-dimethyl-3,4-
dihydro-2H-1-benzopyran (6c)
Yield: 270 mg, 48%. Pale yellow solid. Mp. 107ꢀ108 ꢁC. 1H NMR
4.5.5. 6-(3,4-Dihydroxy-5-methoxy)benzyl-5,7-dihydroxy-2,2-
dimethyl-3,4-dihydro-2H-1-benzopyran (7e)
(600 MHz, CDCl3)
d
10.81 (s, 1H), 7.64 (d, J ¼ 8.8 Hz, 2H), 7.13 (s, 1H),
Yield: 113 mg, 64%. Pale yellow solid. Mp. 95ꢀ97 ꢁC. 1H NMR
6.96e6.92 (m, 2H), 5.86 (s, 1H), 3.85 (s, 3H), 2.60 (t, J ¼ 6.8 Hz, 2H),
1.79 (s, 2H), 1.34 (s, 6H). 13C NMR (151 MHz, CDCl3)
d
196.4, 163.0,
(300 MHz, MeOD)
d
6.88 (dd, J ¼ 10.1, 1.9 Hz, 2H), 5.78 (s, 1H), 3.84
161.4, 160.8, 157.8, 132.0, 130.7, 114.1, 104.0, 101.6, 96.7, 75.9, 55.4,
32.0, 26.7, 16.2. HRMS (ESI) calcd for C19H20O5 [MþH]þ 329.1386
found 329.1384.
(s, 3H), 2.60 (t, J ¼ 6.8 Hz, 2H), 1.80 (t, J ¼ 6.8 Hz, 2H), 1.32 (d,
J ¼ 6.0 Hz, 6H). 13C NMR (75 MHz, MeOD)
d 199.5, 161.3, 161.2, 159.0,
148.8, 145.7, 139.8, 132.8, 112.1, 106.5, 106.0, 101.8, 96.7, 76.4, 56.7,
33.3, 27.0, 17.4. HRMS (ESI) calcd for C19H20O7 [MþH]þ 361.1287
found 361.1277.
4.4.5. 6-(3,4,5-Trimethoxy)benzyl-5,7-dihydroxy-2,2-dimethyl-3,4-
dihydro-2H-1-benzopyran (6d)
Yield: 135 mg, 21%. White solid. Mp. 208ꢀ209 ꢁC. 1H NMR
4.6. Redox potential experiments [29]
(400 MHz, CDCl3) d 11.00 (s, 1H), 7.03 (s, br, 1H), 6.88 (s, 2H), 5.89 (s,
1H), 3.91 (s, 3H), 3.88 (s, 6H), 2.62 (t, J ¼ 6.8 Hz, 2H), 1.81 (t,
The potentials were recorded using a glassy carbon working
electrode, platinum counter electrode and silver wire pseudo
reference electrode. All compounds were measured in CH3CN with
tetrabutyl ammonium tetrafluoroborate as the supporting elec-
trolyte. The solvent was degassed via vigorous argon bubbling prior
J ¼ 6.8 Hz, 2H),1.36 (s, 6H). 13C NMR (101 MHz, CDCl3)
d 196.3,162.2,
161.4, 158.3, 153.9, 141.5, 134.8, 104.9, 103.7, 101.9, 97.0, 76.2, 61.0,
56.4, 32.1, 26.8, 16.2. HRMS (ESI) calcd for C21H24O7 [MꢀH]ꢀ
387.1449 found 387.1455.

Q. Sun et al. / European Journal of Medicinal Chemistry 85 (2014) 621e628
627
to the measurements. All experiments were performed under an
argon atmosphere. Ferrocene was used as the internal reference for
the reduction and oxidation potentials.
antioxidant concentration was calculated for all samples (r2 > 0.93).
Further positive control measurements were performed using
xanthohumol.
4.7. Proliferation assay [30]
Acknowledgements
The proliferation assay was performed using a SV-40T trans-
fected human microvascular endothelial cell line (HMEC-1) [31].
The cells were incubated at 37 ꢁC under a 5% CO2 air atmosphere
with a constant humidity (95%). HMEC-1 were seeded into 96-well
We thank Regina Hoheisel for the redox-potential measure-
ments and Eirini Aplada for providing the picture of Hypericum
empetrifolium Willd. Gabi Brunner is gratefully acknowledged for
providing excellent technical assistance. We are grateful to Josef
microplates (100
m
l, 1.5 ꢂ 103 cells/well) in Endothelial Cell Growth
€
Kiermaier and Wofgang Sollner (Zentrale Analytik NWF IV, Uni
Medium (ECGM) with 10% FCS, Supplement Mix and Antibiotics (all
from Provitro). After 24 h, the medium in a reference plate was
removed, and these cells were stained with a crystal violet solution
for 10 min to obtain a baseline. The cells in the other plates were
treated with increasing concentrations of each test compound
(dissolved in DMSO as a stock solution). After incubating for 72 h,
the cells were stained as previously described. After washing with
Regensburg) for recording the MS spectra and Dr. Ilya Shender-
ovich, Fritz Kastner, Annette Schramm and Georgine Stühler for
measuring the 1D and 2D NMR spectra (Zentrale Analytik NWF IV,
Uni Regensburg). The Chinese Scholarship Council (fellowship to Q.
Sun) is gratefully acknowledged.
Appendix A. Supplementary data
distilled water, 100
absorbance was measured using a Tecan Spectra Fluor Plus at
540 nm. The IC50 values SD were calculated in M using the
ml of a soluble buffer was added and the
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2014.08.017.
m
GraphPad software (from three independent experiments with
different passages and each concentration in hexaplicates). Xan-
thohumol was used as the positive control. Pure ECGM containing
0.1% DMSO was used as the negative control.
References
[1] P. Avato, A Survey on the Hypericum Genus: Secondary Metabolites and
Bioactivity, Elsevier, 2005.
[2] P. Klingauf, T. Beuerle, A. Mellenthin, S.A.M. El-Moghazy, Z. Boubakir,
L. Beerhues, Phytochemistry 66 (2005) 139e145.
4.8. Tube formation assay [30]
[3] a) K. Winkelmann, J. Heilmann, O. Zerbe, T. Rali, O. Sticher, J. Nat. Prod. 64
(2001) 701e706;
b) K. Winkelmann, J. Heilmann, O. Zerbe, T. Rali, O. Sticher, J. Nat. Prod. 63
(2000) 104e108.
[4] a) L. Rocha, A. Marston, O. Potterat, M.A.C. Kaplan, H.S. Evans, K. Hostettmann,
Phytochemistry 40 (1995) 1447e1452;
b) W.K.P. Shiu, S. Gibbons, Phytochemistry 67 (2006) 2568e2572.
[5] a) W. Hashida, N. Tanaka, Y. Kashiwada, M. Sekiya, Y. Ikeshiro, Y. Takaishi,
Phytochemistry 69 (2008) 2225e2230;
HMEC-1 build capillary-like structures on Matrigel™ (BD Bio-
sience, Heidelberg Germany), which is a solubilised basement
membrane extracted from EngelbretheHolmeSwarm (EHS) mouse
sarcoma. After polymerising 10 ml of Matrigel™ in m-slides (ibidi,
Martinsried Germany) within 30 min at 5% CO2, 37 ꢁC, and 95%
humidity, the gel was overlaid with 1 ꢂ 104 HMECs in 25
ml ECGM.
ml ECGM containing 0.1%
The cells were either left untreated (þ25
b) L.H. Lu, K.Y. Sim, Org. Lett. 1 (1999) 879e882;
DMSO) or stimulated with 25 ml of ECGM including four concen-
c) G. Momekov, D. Ferdinandov, D. Zheleva-Dimitrova, P. Nedialkov,
U. Girreser, G. Kitanov, Phytomedicine 15 (2008) 1010e1015.
[6] J. Heilmann, K. Winkelmann, O. Sticher, Planta Med. 69 (2003) 202e206.
[7] V. Butterweck, CNS Drugs 17 (2003) 539e562.
[8] C.M. Schempp, J. Kiss, V. Kirkin, M. Averbeck, B. Simon-Haarhaus, B. Kremer,
C.C. Termeer, J. Sleeman, J.C. Simon, Planta Med. 71 (2005) 999e1004.
[9] B. Martinez-Poveda, A.R. Quesada, M.A. Medina, Int. J. Cancer 117 (2005)
775e780.
trations of the synthesised acylphloroglucinols in their IC50 range.
After 19 h, the cells were photographed using a PrimoVert micro-
scope (Zeiss, Oberkochen Germany). Each experiment was per-
formed in triplicate. One representative picture is shown.
4.9. ORAC-fluorescein assay
[10] C.Y.W. Ang, L.H. Hu, T.M. Heinze, Y.Y. Cui, J.P. Freeman, K. Kozak, W.H. Luo,
J. Agric. Food Chem. 52 (2004) 6156e6164.
[11] S. Schmidt, G. Jürgenliemk, H. Skaltsa, J. Heilmann, Phytochemistry 77 (2012)
218e225.
[12] S. Schmidt, G. Jürgenliemk, T.J. Schmidt, H. Skaltsa, J. Heilmann, J. Nat. Prod. 75
(2012) 1697e1705.
[13] K. Winkelmann, M. San, Z. Kypriotakis, H. Skaltsa, B. Bosilij, J. Heilmann,
Z. Naturforschung C J. Biosci. 58 (2003) 527e532.
[14] Y. Ishida, O. Shirota, S. Sekita, K. Someya, F. Tokita, T. Nakane, M. Kuroyanagi,
Chem. Pharm. Bull. 58 (2010) 336e343.
[15] H. Jayasuriya, J.D. Mcchesney, J. Chem. Soc. Chem. Commun. (1988)
1592e1593.
[16] S.G. Cao, J.K. Schilling, J.S. Miller, R. Andriantsiferana, V.E. Rasamison,
D.G.I. Kingston, J. Nat. Prod. 67 (2004) 454e456.
[17] A. Davalos, C. Gomez-Cordoves, B. Bartolome, J. Agric. Food Chem. 52 (2004)
48e54.
[18] F.M. Roleira, C. Siquet, E. Orrù, E.M. Garrido, J. Garrido, N. Milhazes, G. Podda,
F. Paiva-Martins, S. Reis, R.A. Carvalho, E.J. Silva, F. Borges, Bioorg. Med. Chem.
18 (2010) 5816e5825.
[19] G.P. Kalena, A. Jain, A. Banerji, Molecules 2 (1997) 100e105.
[20] R.H. Cichewicz, V.A. Kenyon, S. Whitman, N.M. Morales, J.F. Arguello,
T.R. Holman, P. Crews, J. Am. Chem. Soc. 126 (2004) 14910e14920.
[21] L. Crombie, R.C.F. Jones, C.J. Palmer, J. Chem. Soc. Perkin Trans. 1 (1987)
317e331.
[22] B.L. Booth, G.F.M. Noori, J. Chem. Soc. Perkin Trans. 1 (1980) 2894e2900.
[23] C.M. Lin, S.T. Huang, F.W. Lee, H.S. Kuo, M.H. Lin, Bioorg. Med. Chem. 14 (2006)
4402e4409.
The antioxidant activity was determined via an oxygen radical
absorbance capacity-fluorescein (ORAC-FL) assay [16,17] as previ-
ously described [32,33]. In brief, the ORAC-FL assay was performed
in a 96-well plate containing fluorescein (final concentration
300 nM) as a fluorescent probe and using a 75 mM phosphate
buffer (pH 7.4) for all dilution steps and as a reaction milieu. The
antioxidant (test compounds or Trolox, 20
different concentrations (test compounds: 1e5
1e8
M) with a fluorescein solution (120 mL) at 37 ꢁC for 15 min.
The reaction began upon the addition of 60 L AAPH (2,20-azobis-
(2-methylpropionamide)-dihydrochloride, final concentration:
12 mM) for a final volume of 200 L. After adding AAPH, the fluo-
mL) was incubated to
mM, Trolox:
m
m
m
rescence was recorded every minute in a Tecan 96-plate reader (lex
485 nm, lem 536 nm, 37 ꢁC) for 80 min. The reaction mixtures were
prepared in quadruplicate, and at least four independent assays
were performed for each sample. The samples were measured at
five different concentrations (1e5
mM). Eight calibration curves
were obtained for each assay using 1e8
mM Trolox as the antioxi-
dant. The controls were measured without an antioxidant or
without AAPH and antioxidant. The ORAC values were expressed as
Trolox equivalents (mean SD) using the standard curve calculated
for each assay. The Regression coefficient between the AUC and
[24] S. Mizobuchi, Y. Sato, Agric. Biol. Chem. 49 (1985) 719e724.
[25] J.H. George, M.D. Hesse, J.E. Baldwin, R.M. Adlington, Org. Lett. 12 (2010)
3532e3535.
[26] Y.R. Lee, X. Li, J.H. Kim, J. Org. Chem. 73 (2008) 4313e4316.

628
Q. Sun et al. / European Journal of Medicinal Chemistry 85 (2014) 621e628
[27] Q.L. Wang, X.G. She, X.F. Ren, J.Y. Ma, X.F. Pan, Tetrahedron Asymmetry 15
(2004) 29e34.
[28] F.L. Zheng, S.R. Ban, X.E. Feng, C.X. Zhao, W.H. Lin, Q.S. Li, Molecules 16 (2011)
4897e4911.
[31] E.W. Ades, F.J. Candal, R.A. Swerlick, V.G. George, S. Summers, D.C. Bosse,
T.J. Lawley, J. Investig. Dermatol. 99 (1992) 683e690.
[32] a) S. Vogel, S. Ohmayer, G. Brunner, J. Heilmann, Bioorg. Med. Chem. 16 (2008)
4286e4293;
[29] R.R. Gagne, C.A. Koval, G.C. Lisensky, Inorg. Chem. 19 (1980) 2854e2855.
[30] A. Koltermann, Influence of Ginkgo Biloba Extract EGb 761 on Signaling
Pathways in Endothelial Cells, LMU, München, 2008.
b) S. Vogel, M. Barbic, G. Jürgenliemk, J. Heilmann, Eur. J. Med. Chem. 45
(2010) 2206e2213.
[33] M. Decker, B. Kraus, J. Heilmann, Bioorg. Med. Chem. 16 (2008) 4252e4261.